Prediction Of The Dew-point Pressure Of A Gas Condensate Reservoir

This study was carried out to predict the dew point pressure of a gas condensate reservoir. The Dew point pressure (DPP) is one amongst the foremost essential parameters to characterize gas condensate reservoirs. Experimental determination of DPP during a window pressure-volume-temperature cell is commonly tough particularly just in case of lean retrograde gas condensation. Therefore, checking out quick and sturdy algorithms for determination of DPP is sometimes required.

This Paper presents a replacement approach supported artificial neural network (ANN) to work out DPP for gas condensation reservoirs specifically within the Niger delta. Therefore, it addresses the inequality of already existing correlation within the prediction of dew point pressure. In developing the model during this work, five hundred data sets from the Constant volume depletion experiment done from gas condensates taken from fields within the Niger Delta were used. The data sets were indiscriminately divided into four sets. For coaching, validation, testing the Neural network developed. The feed forward back-propagation learning algorithms were employed in the network because it became the best approach. The neural design used, concerned two hidden layers that ultimately improved the results of the network.

The accuracy of this developed model was tested with fifty data samples and also the result was compared to the results from already existing correlations and Neural network model developed. It was absolutely and clearly seen that the projected model from this work performed higher with a parametric statistic of 0.99999 and AARE of 0.039675 than alternative correlations and existing models